One-third of drivers convicted of DUI (driving under the influence of alcohol) will repeat the offense. These drivers present a significantly elevated risk to the public and to themselves. The majority of first-time DUI offenders are assigned to a short-term education program as part of their court sanction or as a requirement for license reinstatement; however, with such high recidivism rates, it is evident that these programs can be improved. Whereas traditional programs attempt to motivate offenders to control their drinking in order to avoid a future DUI arrest, a new program-PARC (Preventing Alcohol-Related Convictions)-introduces a strategy of controlling driving (to drinking events) rather than controlling drinking (at drinking events). This harm reduction approach grows out of research on the reasons DUI offenders fail to avoid additional arrests for impaired driving. These researchers found that offenders who drove to a drinking event would drive home impaired because, despite their intention to avoid drinking, they could not control their drinking in environments where others were consuming alcohol. Although the PARC course contains the standard information on alcohol consumption and alcohol use disorders that provide a basis for controlling drinking, the central theme of the PARC curriculum is that leaving one's vehicle at home is the most effective way to prevent driving while impaired. An opportunity to initiate a statewide random clinical trial of the PARC curriculum and its relatively simple protective concept was provided by the State of Florida, Department of Highway Safety and Motor Vehicles. Between August 2004 and March 2006, with SBIR funding from the NIAAA, the PARC and the Traditional first offender curricula were randomly assigned to 43,000 first DUI offenders in Florida. An evaluation of the first 10,000 participants with a minimum of one year of driving exposure showed the PARC curriculum was associated with a 42% reduction in recidivism. This application proposes a secondary analysis of the records of the full 43,000 participants that can now be tracked for 2 to 4 years after program completion to verify the benefits shown in the preliminary study and to determine which type of offender benefits the most and which benefits the least (i.e., males, underage drivers, heavy drinkers, Hispanics, and others), thereby, providing crucial information about the continuation, expansion, or modification of the PARC program. [unreadable] [unreadable] [unreadable]